Improvement of Transplanting Rice Yield and Nitrogen Use Efficiency by Increasing Planting Density in Northeast China Under the Optimal Nitrogen Split-Fertilizer Applications

Abstract

There are few studies on how nitrogen (N) fertilizer application rates and transplanting densities impact rice yield, root distribution, and N use efficiency in the cold regions of Northeast China. This research involved a two-year field trial utilizing Jinongda 667 as the material. In 2021, three N split-fertilizer applications—T1 (6:3:1), T2 (5:3:2), T3 (4:3:3)—and two transplanting densities—D1 (30 cm × 13.3 cm) and D2 (30 cm × 20 cm)—were compared with the conventional cultivation mode (T0: 175 kg N hm⁻², 6:3:1), whereby the N application mode most suitable for increasing density was explored. In 2022, four N application levels—0 (N0), 125 (N1), 150 (N2), and 175 (N3) kg N hm⁻²—were assessed under the same density treatment to analyze the yield, resource utilization efficiency, and root traits of Jinongda 667. The results indicated that when the transplanting density was 30 cm × 13.3 cm, the application of 5:3:2 fertilizer was more conducive to improving rice yield. Increasing planting density under reduced N input significantly enhanced both rice yield and N use efficiency. In contrast to the conventional cultivation method (D2N3), the treatment of increased planting density (D1N2) under reduced N input led to a 21.2% rise in the number of panicles per square meter and an 8.6% boost in rice yield. Furthermore, increasing planting density under reduced N input significantly enhanced the agronomic efficiency of N fertilizer, the apparent utilization rate, and the N harvest index. It also boosted the SPAD value, photosynthetic rate, and the utilization efficiency of light and N resources in rice. However, it was noted that root enzyme activity decreased. This study demonstrated that increasing planting density, combined with the N application mode of 5:3:2 and an N application rate of 150 kg hm⁻², maximized resource utilization efficiency, optimized root absorption capacity, and resulted in higher yields.

1. Introduction

Rice (Oryza sativa L.) is a staple crop in China, and its high yield is crucial for the country’s food security. Currently, the average N application rate for rice in China is 180 kg hm⁻², which is 75% higher than the global average [1,2]. Historically, farmers have often obtained high yields by applying large amounts of fertilizer. While the excessive utilization of N fertilizer has led to a considerable enhancement in yield, it has concurrently resulted in a reduction in nitrogen use efficiency and the emergence of significant environmental contamination [3,4,5]. In addition, overfertilization can cause rice greed, leading to a prolonged vegetative growth period that consumes more nutrients. This results in insufficient N required for the reproductive growth period, poor grain filling, lower seed-setting rates, increased risk of lodging, and higher susceptibility to pests and diseases, ultimately reducing yield [6]. Therefore, ensuring increased rice yields while minimizing environmental impact is essential for achieving green, sustainable rice production in China [7].

Optimizing fertilizer application and raising planting density have been demonstrated to considerably improve both rice yield and N use efficiency. Liu et al. [8] demonstrated that an optimal N application strategy with a 4:3:3 ratio, applied before transplanting, at the tillering stage and at the heading stage, can significantly improve N use efficiency. This approach also boosts yield in late mechanically transplanted rice. Chong et al. [9] found that the yield loss caused by reduced N application can be offset by appropriately increasing planting density. An optimal planting density for rice can effectively balance the competition between plants within groups and for individuals, and can help regulate yield components. Zhou et al. [10] observed that increasing planting density significantly reduces humidity between plants, which lowers the incidence of pests and diseases, slows down leaf area senescence in the late growth stage, and ultimately enhances the photosynthetic function and material production capacity of rice. Furthermore, various research efforts indicate that decreasing N input while raising planting density may enhance nitrogen use efficiency in rice by 23.3% to 31.9%, along with improving nitrogen recovery efficiency by 17.4% to 24.1% [11,12,13]. Moreover, studies have shown that [14,15] increasing planting density with reduced N input can optimize canopy light capture and N structure, promoting improved radiation use efficiency (RUE) and N use efficiency for grain production (NUEg). This approach enhances dry matter production and distribution, leading to a synergistic increase in both yield and resource utilization efficiency. However, excessively high planting density often results in intense competition among rice plants, limiting individual growth of rice, strengthening overall plant density excessively, and reducing the number of grains per spike, which is counterproductive to achieving high yields. Therefore, coordinating N application rates with planting density is essential for high-yield, high-efficiency rice production [16]. Generally, enhancing rice yield by optimizing cultivation methods, such as basic seedling density and N fertilizer management, to achieve optimal cultivation outcomes is the preferred approach today [17,18].

Current studies on the effects of increasing density and reducing N levels on crop yield and quality are mainly focused on the southern region [19,20,21]. However, there have been few studies conducted in the cold regions of Northeast China. To address this gap, we explored the impacts of N application rate and transplanting density on the yield and N use efficiency of japonica rice in northern China through field experiments. We additionally chose the best combination of the N application, N application method, and the interaction treatment of transplanting density to provide a theoretical foundation and technical support for developing a high-yield, high-efficiency rice cultivation model in Central Jilin Province.

2. Materials and Methods

2.1. Experimental Site and Soil Properties

The research was carried out in the years 2021 and 2022 at the Characterization Station for National Crop Variety Approval at Jilin Agricultural University, located in Changchun City, Jilin Province. For this study, we selected Jinongda 667, an exemplary high-quality edible rice variety known for its significance in the region of Jilin Province, as the test variety. The physical and chemical properties of the soil used in the experiment for both years have been detailed in

Table 1. Soil physico-chemical characteristics in 2021 and 2022.

Year	Organic Matter (g kg−1)	Alkaline Dissolved Nitrogen (mg kg−1)	Available Phosphorus (mg kg−1)	Available Potassium (mg kg−1)	PH
2021	16.75	26.53	17.78	137.09	6.7
2022	14.20	33.89	29.42	137.09	6.7

.

2.2. Experimental Design

In this study, a completely randomized block design was used to conduct three repeated experiments. In 2021, the total application rates for nitrogen, phosphorus, and potassium were set at 150 kg hm⁻², 90 kg hm⁻², and 105 kg hm⁻², respectively. The ratios of base fertilizer, tillering fertilizer, and panicle fertilizer were established as follows: T1 (6:3:1), T2 (5:3:2), and T3 (4:3:3). The conventional fertilization (T0) was used as the control, with an N fertilizer application of 175 kg hm⁻². The ratio of base fertilizer, tillering fertilizer, and panicle fertilizer was set at 6:3:1. Two transplanting densities were also set: D1 (30 cm × 20 cm) and D2 (30 cm × 13.3 cm), resulting in a total of eight treatments, each repeated three times. The plot area for each treatment was 30 m². In 2022, to further investigate the impacts of increasing density and reducing N application on rice yield, root traits, and N use efficiency under optimal fertilizer operation (with a base fertilizer, tillering fertilizer, and panicle fertilizer ratio of 5:3:2), four N application rates were set up: no N application (N0), N application rate of 125 kg hm⁻² (N1), N application rate of 150 kg hm⁻² (N2), and the farmers’ traditional N application of 175 kg hm⁻² (N3). At the same time, two rice transplanting densities were set: D1 (30 cm × 20 cm) and D2 (30 cm × 13.3 cm), resulting in a total of 8 treatments. Each plot area was 30 m². In 2022, seedlings were raised on 5 April, transplanting was completed by 20 May, and the harvest and yield measurements were taken on 28 September. In addition, seedlings were raised on 13 April and artificially transplanted on 25 May. The seedlings had an average of 3 leaves and 1 heart, with 5 plants per hole, and were harvested on 30 September. Each experimental plot could be irrigated and drained separately, and other management practices were conducted according to local conventional high-yield cultivation methods.

2.3. Sampling and Measurements

2.3.1. Grain Yield and Yield Components

At the point of maturity, five representative points of 1 m² were selected and the number of panicles per m² was calculated. Subsequently, 10 hills with an identical number of panicles were selected from the five points for drying. Thereafter, the spikelets per panicle, 1000-grain weight, and filled grain rate were calculated. Furthermore, five random sampling points were selected within each plot, excluding edge rows and initial sampling rows. Furthermore, a total area of 3 m² was harvested at each point in order to assess the yield per unit area. Subsequently, the moisture content of the rice grains was determined for each plot following the completion of the drying process. The final grain yield was calculated by adjusting the obtained data to a standard moisture content of 13.5%.

2.3.2. Nitrogen Use Efficiency

N accumulation and N use efficiency: Sampling was conducted at the maturity stage of rice. Five plants were selected from each plot and separated into stems, leaves, and panicles. The samples were deactivated at 105 °C for 30 min and then dried at 80 °C until they reached a constant weight. After grinding, the samples were digested using H₂SO₄-H₂O₂, and the N content of each organ sample was determined using an automatic Kjeldahl nitrogen analyzer. The calculation formula is as follows:

Total N uptake (kg hm⁻²) = above-ground dry weight at maturity × N content

Agronomic use efficiency of N (AEN) (kg/kg) = (yield in N application area-yield in N-free area)/N application rate

Recovery efficiency of N (REN) (%) = (total N uptake of plants in N application area–total N uptake of plants in N-free area)/N application rate × 100%

N fertilizer productivity bias (NFP) (kg/kg) = yield in N application area/N application rate

N harvest index (NHI) (%) = grain N accumulation at harvest/total plant N accumulation

Physiological N use efficiency (PNUE) (kg/kg) = (yield in N application area–yield in N-free area)/(total N uptake in N application area–total N uptake in N-free area)

2.3.3. SPAD Value

During the mid-tillering (MT), panicle initiation (PI), heading (HD), and filling stages (FS), measurements of SPAD values were conducted on the flag leaves of rice. In each plot, five points were chosen (excluding the edge rows), and at every point, five rice plants with comparable growth characteristics were selected based on the average tiller count. The SPAD values, which serve as an indirect measure of chlorophyll content in the leaves, were recorded using a SPAD-502 portable chlorophyll meter (Minolta, Osaka, Japan) to determine the chlorophyll levels in the flag leaves. Each measurement was taken five times to ensure accuracy.

2.3.4. Root Traits

Soil blocks measuring 20 cm × 20 cm × 20 cm were dug from the root stubble of each hole at the MT, PI, HD, and FS, respectively, with 5 holes sampled for each treatment. The samples were placed in a bucket, and the roots were rinsed with running water. After the above-ground portion had been removed, the volume of the roots was measured by the drainage method. Some roots were dried and weighed to determine the root dry weight.

2.3.5. Root Activity

During the specified period, 5 holes exhibiting relatively consistent growth were selected. The rice was harvested using the abovementioned method and placed back into the barrel, with the roots carefully washed with clear water. Samples of 0.5 g of root tips were weighed and placed in a small beaker, to which 5 mL of 0.4% TTC solution and 5 mL of phosphate buffer (pH 7.0) were added, ensuring that the roots were fully immersed in the solution. The root samples were maintained in the dark at 37 °C for a period of between one and two hours, after which 2 mL of 1 mol/L sulfuric acid was immediately added to terminate the reaction. At the same time, a blank experiment was conducted. In this case, sulfuric acid was added first, followed by the root sample. The apparatus was maintained at a temperature of 37 °C in the absence of sulfuric acid. The procedure for determining the degree of the solution and the operational steps were identical to those previously described. Subsequently, the roots were excised, desiccated with filter paper, and placed in a mortar. A quantity of 3–4 mL of ethyl acetate was then added, and the mixture was subjected to thorough grinding in order to facilitate the extraction of TTF. The red extract was then transferred to a graduated test tube. Subsequently, the residue was washed with a minimal quantity of ethyl acetate on two or three occasions, after which all the samples were weighed. Subsequently, ethyl acetate was added to achieve a total volume of 10 mL. A spectrophotometer was used to measure the absorbance at a wavelength of 485 nm. The standard curve was used as a reference point for determining the quantity of TTC that had been reduced [22].

2.3.6. Root Enzyme Activity

Five holes with consistent growth vigor were sampled at the MT, HD, and FS stages, respectively. After harvesting the rice using this method, the roots were immediately cleaned with distilled water and dried with filter paper. A small section of fresh roots, 2–4 cm above the root tip, was wrapped in tin foil and placed back in liquid nitrogen. The BC0080 testing kits provided by Beijing Solaibao Technology Co., Ltd. (located in Beijing, China) were employed to assess the levels of nitrate reductase (NR) via UV spectro-photometric analysis. Additionally, the BC0910 kit was utilized independently to measure the activity of glutamine synthase (GS) through visible spectrophotometry.

2.3.7. Statistical Analysis

Microsoft Excel 2019 and SPSS 21 software were used for data processing. In addition, the least significant difference (LSD) method was applied for multiple comparisons. Drawing was carried out using Origin 2024.

3. Results

3.1. Yield and Yield Components

3.1.1. Yield and Yield Components in 2021

The results in

Table 2. Yield and yield components for Jinongda 667 under different N fertilizer management ratio and planting density treatments in 2021.

Density	Nitrogen Fertilizer Rate	Panicle Number (m−2)	Spikelets Panicle−1	Grain Filling (%)	Grain Weight (g/1000 Seeds)	Grain Yield (t hm−2)
D1	T0	362.23 ± 4.62 bc	135.45 ± 2.91 ab	91.30 ± 1.70 cd	22.18 ± 0.36 ab	8.97 ± 0.15 ab
	T1	369.51 ± 16.50 bc	134.24 ± 1.67 ab	92.63 ± 0.88 bcd	21.35 ± 0.64 b	8.39 ± 0.32 bc
	T2	353.23 ± 9.24 c	138.41 ± 1.95 ab	93.57 ± 0.35 ab	22.05 ± 0.93 ab	8.81 ± 0.10 abc
	T3	339.42 ± 9.81 c	144.27 ± 3.08 a	94.82 ± 2.09 a	22.72 ± 0.67 a	9.06 ± 0.10 ab
D2	T0	413.60 ± 12.50 a	131.21 ± 7.04 bc	88.65 ± 0.96 d	22.18 ± 0.44 ab	9.11 ± 0.31 ab
	T1	406.23 ± 14.23 a	121.35 ± 3.28 d	91.34 ± 1.62 bcd	22.48 ± 0.13 a	8.76 ± 1.04 bc
	T2	417.54 ± 14.16 a	125.86 ± 4.79 cd	92.81 ± 1.10 abc	22.94 ± 0.40 a	9.51 ± 0.06 a
	T3	386.47 ± 14.23 b	137.53 ± 1.45 a	93.57 ± 0.54 abc	21.78 ± 0.37 ab	8.21 ± 0.17 c
Analysis of variance
D		**	**	NS	NS	NS
R		*	**	**	NS	NS
D × R		NS	*	NS	*	*

Note: Different letters after columns indicate statistical significance at the p < 0.05 level between treatments of the same density. Significant differences in ANOVA at the 0.05 and 0.01 levels are indicated by * and **, respectively, and NS indicates non-significant differences. D1, 30 cm × 20 cm; D2, 30 cm × 13.3 cm; T0, 6:3:1 175 kg N hm⁻²; T1, 6:3:1 175 kg N hm⁻²; T2, 5:3:2 175 kg N hm⁻²; T3, 4:3:3 175 kg N hm⁻².

show that the T3 treatment yielded the most under D1 density. In contrast, the T2 treatment yielded the most under D2 density. Specifically, the yield of the D2T2 treatment was 4.9% higher than that of the D1T3 treatment. Compared to D1T3 treatment, D2T2 treatment had a higher effective panicle number and grain weight. Additionally, the yield of D2T2 was 4.4% greater than that of D2T0, largely due to the higher seed setting rate. In comparison to the D1T0 treatment, the yield increased by 6.0%, mainly due to an increase in the effective number of spikes per unit area. This indicates that a transplanting density of 30 cm × 13.3 cm, combined with a fertilizer management ratio of 5:3:2, is more effective for improving rice yield.

3.1.2. Yield and Yield Components in 2022

Table 3. Yield and yield components for Jinongda 667 under different N and planting density treatments in 2022.

Density	Nitrogen	Panicles Number (m−2)	Spikelets Panicle−1	Grain Filling (%)	Grain Weight (g/1000 Seeds)	Grain Yield/ (t hm−2)
D1	N0	129.2 ± 10.46 f	167.80 ± 22.1 cd	97.55 ± 1.53 a	24.60 ± 0.52 a	4.47 ± 0.34 d
	N1	258.4 ± 14.21 d	184.7313.02 bc	96.89 ± 1.64 a	23.02 ± 0.58 b	8.80 ± 0.11 c
	N2	292.4 ± 14.22 cd	190.90 ± 18.22 ab	96.70 ± 1.82 a	22.50 ± 1.40 b	9.49 ± 0.17 b
	N3	299.2 ± 32.14 bcd	212.80 ± 18.74 a	96.26 ± 0.95 a	21.94 ± 0.71 b	9.50 ± 0.15 b
D2	N0	215.0 ± 9.14 e	152.94 ± 15.09 d	97.50 ± 0.86 a	23.28 ± 0.61 ab	4.85 ± 0.29 d
	N1	305.0 ± 16.14 bc	199.14 ± 11.53 ab	96.83 ± 2.02 a	21.75 ± 0.73 b	9.21 ± 0.30 bc
	N2	362.5 ± 18.41 a	202.50 ± 13.16 ab	95.99 ± 2.66 a	21.79 ± 0.33 b	10.32 ± 0.26 a
	N3	335.0 ± 13.78 ab	197.47 ± 16.77 ab	95.34 ± 1.35 a	22.80 ± 1.10 b	9.70 ± 0.34 ab
Analysis of variance
D		**	NS	NS	NS	**
N		**	**	**	**	**
D × N		NS	NS	NS	NS	NS

Note: Different letters after columns indicate statistical significance at the p < 0.05 level between treatments of the same density. Significant differences in ANOVA at the 0.01 levels is indicated by **, and NS indicates non-significant differences. D1, 30 cm × 20 cm; D2, 30 cm × 13.3 cm; T0, 6:3:1 175 kg N hm⁻²; T1, 6:3:1 175 kg N hm⁻²; T2, 5:3:2 175 kg N hm⁻²; T3, 4:3:3 175 kg N hm⁻².

shows that the maximum yield for the N3 treatment at D1 density was 9.5 t hm⁻². In contrast, the maximum yield for the N3 treatment at D2 density was 10.3 t hm⁻². The yield for the D2N2 treatment was 8.6% higher than that for the D1N3 treatment, mainly due to a 21.2% increase in the number of panicles. This indicates that reducing N application to 150 t hm⁻² and increasing density are beneficial to increasing rice yield.

3.2. N Use Efficiency

Table 4. Effect of interaction between nitrogen application rate and planting density on rice nitrogen use efficiency.

Density	Nitrogen Rate	Total Nitrogen Accumulation	AEN	REN	NFP	NHI	PNUE
		(kg hm−2)	(kg kg−1)	(%)	(kg kg−1)	(%)	(kg kg−1)
D1	N0	72.71 ± 6.35 e
	N1	135.21 ± 9.36 c	28.10 ± 2.54 b	50.00 ± 0.88 c	70.37 ± 6.28 a	69.56 ± 3.75 b	50.59 ± 1.34 a
	N2	166.78 ± 9.77 b	30.56 ± 1.15 bc	62.71 ± 1.15 b	63.29 ± 3.45 b	71.64 ± 4.96 ab	45.07 ± 1.55 bc
	N3	180.70 ± 5.51 b	28.76 ± 0.85 d	61.71 ± 0.85 bc	54.30 ± 3.15 c	72.81 ± 2.36 ab	39.24 ± 1.16 d
D2	N0	91.35 ± 7.99 d
	N1	166.55 ± 7.24 b	34.91 ± 3.00 b	60.16 ± 2.56 bc	72.80 ± 4.58 a	72.41 ± 4.92 ab	51.97 ± 4.10 a
	N2	209.31 ± 8.35 a	39.02 ± 2.78 a	78.64 ± 1.60 a	70.13 ± 5.57 a	74.77 ± 2.66 a	48.97 ± 3.47 ab
	N3	208.41 ± 10.80 a	29.88 ± 1.92 cd	66.89 ± 1.92 b	55.42 ± 2.16 c	70.46 ± 1.84 b	40.53 ± 2.59 cd
Analysis of variance
D		**	NS	NS	**	NS	NS
N		**	**	NS	**	NS	**
D × N		NS	NS	NS	*	NS	NS

Note: AEN, Agronomic use efficiency of N; REN, Recovery efficiency of N; NFP, N fertilizer productivity bias; NHI, N harvest index; PNUE, Physiological N use efficiency. Different letters after columns indicate statistical significance at the p < 0.05 level between treatments of the same density. Significant differences in ANOVA at the 0.05 and 0.01 levels are indicated by * and **, respectively, and NS indicates non-significant differences. D1, 30 cm × 20 cm; D2, 30 cm × 13.3 cm; T0, 6:3:1 175 kg N hm⁻²; T1, 6:3:1 175 kg N hm⁻²; T2, 5:3:2 175 kg N hm⁻²; T3, 4:3:3 175 kg N hm⁻².

shows that the total nitrogen (N) accumulation of D2N2 was 15.8% higher than that of D1N3. In addition, the AEN, REN, and NFP of D2N2 were significantly higher than those of D1N3, at 35.7%, 27.4%, and 29.2%, respectively. The total N accumulation, AEN, REN, and NHI of each treatment were the highest in the D2N2 treatment. Furthermore, the AEN and REN values for the D2N2 treatment were notably higher compared to the other treatments. Meanwhile, NFP and PNUE were highest in the D2N1 treatment. These results indicate that rice grown with a reduced N application rate alongside an increased planting density (D2N2) can enhance its N use efficiency.

3.3. SPAD Value

As shown in Figure 1, the SPAD value of rice leaves in each treatment increased gradually at the MT, PI, and HD stages, peaking before beginning to decrease at the FS stage. Under the normal density treatment, the SPAD value generally increased with the nitrogen application rate. For different density treatments, increasing planting density raised SPAD values of leaf blades at both the HD and FS stages. The SPAD value of rice leaves was highest in the D2N2 treatment at HD and FS. This indicates that increasing planting density with reduced N input allows rice to maintain a high SPAD value during the late growth stage.

3.4. Net Photosynthetic Rate

Table 5. Effect of interaction between nitrogen application rate and planting density on rice Pn.

Density	Nitrogen	Pn (μmol·m−2 s−1)
		MT	PI	HD	FS
D1	N1	20.23 ± 0.76 a	20.15 ± 1.47 a	17.73 ± 0.49 c	13.23 ± 1.04 b
	N2	20.29 ± 0.21 a	21.08 ± 1.66 a	19.41 ± 0.97 b	14.59 ± 0.29 a
	N3	21.02 ± 0.33 a	21.27 ± 0.35 a	20.00 ± 0.72 ab	14.75 ± 0.21 a
D2	N1	20.31 ± 0.85 a	20.41 ± 1.12 a	18.19 ± 0.52 c	13.64 ± 0.39 b
	N2	20.75 ± 0.93 a	22.06 ± 1.12 a	20.67 ± 0.33 a	15.65 ± 0.39 a
	N3	21.07 ± 0.51 a	21.53 ± 1.43 a	20.80 ± 0.62 a	15.41 ± 0.32 a
Analysis of variance
D		NS	NS	**	**
N		NS	NS	*	*
D × N		NS	NS	NS	NS

Note: Pn, Net photosynthetic rate; MT, Mid-tillering; PI, Panicle initiation; HD, Heading; FS, filling stage. Different letters after columns indicate statistical significance at the p < 0.05 level between treatments of the same density. Significant differences in ANOVA at the 0.05 and 0.01 levels are indicated by * and **, respectively, and NS indicates non-significant differences. D1, 30 cm × 20 cm; D2, 30 cm × 13.3 cm; T0, 6:3:1 175 kg N hm⁻²; T1, 6:3:1 175 kg N hm⁻²; T2, 5:3:2 175 kg N hm⁻²; T3, 4:3:3 175 kg N hm⁻².

shows that the net photosynthetic rate of D2N3 was highest at both the PI and HD stages. In contrast, D2N2 exhibited the highest rate at the PI and FS stages. Compared to the D1N3 treatment, the photosynthetic rates under the D2N2 treatment increased by 2.7%, 3.4%, and 6.5% at PI, HD, and FS, respectively. This indicates that reducing the rate of N application and increasing planting density (D2N2) can enhance the net photosynthetic rate. In addition, it can slow the decline of this rate during the late growth stage.

3.5. Root Traits

3.5.1. Root Physical Properties

Figure 2 shows that the root length, root volume, and root dry weight of rice at each key growth stage were significantly influenced by N fertilizer treatment. In addition, the interaction between N fertilizer and planting density also had a notable impact on these root characteristics. The D2N3 treatment yielded the highest root length, root volume, and root dry weight at the MT stage. In contrast, the D2N2 treatment produced the highest values at the PI, HD, and FS stages. Compared to the D1N3 treatment, the D2N2 treatment showed an increase in root length of 2.0% to 8.3%, root volume of 2.5% to 6.6%, and root dry weight of 5.06% to 16.62% from the booting stage to the flowering stage (FS). When planting density was increased under reduced N input, significant increases in root volume and dry weight were observed at the PI, HD, and FS stages compared to the traditional N density treatment. Therefore, reducing the amount of N applied while increasing planting density resulted in improved root physical properties in rice.

3.5.2. Root Activity

Table 6. Effect of interaction between nitrogen application rate and planting density on rice root viability.

Density	Nitrogen Rate	Root Activity (μg·g−1 h−1)
		MT	PI	HD	FS
D1	N1	116.20 ± 8.91 b	191.55 ± 10.38 c	127.99 ± 11.03 d	108.89 ± 2.58 b
	N2	133.95 ± 16.24 ab	221.48 ± 14.31 bc	147.00 ± 7.97 c	132.21 ± 4.25 a
	N3	145.78 ± 8.58 a	242.88 ± 16.76 ab	150.96 ± 5.77 c	110.45 ± 7.55 b
D2	N1	123.85 ± 8.65 ab	220.09 ± 14.11 bc	159.01 ± 6.53 bc	60.86 ± 8.88 d
	N2	139.34 ± 10.31 ab	266.55 ± 12.10 a	184.59 ± 7.33 a	77.74 ± 10.28 c
	N3	146.82 ± 7.01 a	246.19 ± 13.93 ab	166.49 ± 6.26 b	84.53 ± 10.27 c
Analysis of variance
D		NS	NS	NS	NS
N		**	NS	NS	NS
D × N		NS	NS	NS	NS

Note: Different letters after columns indicate statistical significance at the p < 0.05 level between treatments of the same density. Significant differences in ANOVA at the 0.01 levels are indicated by **, and NS indicates non-significant differences. MT, Mid-tillering; PI, Panicle initiation; HD, Heading; FS, filling stage. D1, 30 cm × 20 cm; D2, 30 cm × 13.3 cm; T0, 6:3:1 175 kg N hm⁻²; T1, 6:3:1 175 kg N hm⁻²; T2, 5:3:2 175 kg N hm⁻²; T3, 4:3:3 175 kg N hm⁻².

shows that the root activity of rice was highest in D2N3 at MT, highest in D2N2 at PI and HD, and highest in D1N2 at FS. Among these, D2N2 demonstrated root activity that was 10.9%, 22.3%, and 25.6% higher than D2N3, D1N3, and D1N2 at HD, and 19.7%, 56.4%, and 70.1% higher at FS, respectively. The root activity of rice in each treatment peaked at PI, with the decrease rates for D2N2 and D2N1 at HD and PI being lower than those for D1N2 and D1N1. This indicates that appropriately reducing the amount of N application and increasing planting density can enhance root activity in rice and maintain it during the later stages of growth.

3.5.3. Enzyme Activity of Roots

From Figure 3, it can be observed that the NR activity of rice roots in each treatment exhibited a gradually decreasing trend as the growth period advanced. The activity of GS initially increased and then decreased over the same period. Regarding N treatment, both NR and GS activity increased with higher N application rates under the same density treatment. In terms of density treatment, there was no significant difference in NR and GS activity between the increased planting density and normal density treatments. Compared to the D1N3 treatment, the NR activity in the D2N2 treatment decreased during each growth period. However, the D2N2 treatment still maintained higher GS activity than the D1N3 treatment during the HD and FS periods. Overall, increasing planting density while reducing N input led to decreased NR and GS activity in rice roots.

4. Discussion

4.1. Effects of Different Planting Density, N Management Mode, and N Application Rate on Yield and Yield Components

Transplanting density is an effective cultivation method for regulating rice population, significantly impacting rice yield [23]. In this study, when the density increased from 30 cm × 20 cm to 30 cm × 13.3 cm, the yield and number of panicles m⁻² increased by 3.0% and 18.7%, respectively, compared to the 30 cm × 20 cm density. The increase in planting density mainly increased the number of panicles m⁻² of rice, which in turn affected the overall yield [24]. Previous studies have also indicated that increasing planting density mainly boosts rice yield by increasing the number of panicles m⁻² [20]. In addition, rice yield has been shown to be significantly influenced by the interaction between N fertilizer and density [25], which aligns with our results.

The results of previous studies on the impacts of fertilizer management on rice yield vary. N application methods significantly impact rice yield and its composition. An appropriate fertilization ratio can effectively increase the effective biomass of the rice population, build a population with high photosynthetic efficiency, and significantly increase the number of spikelets, resulting in higher panicles m⁻² and spikelets panicle⁻¹. This approach also enhances the material productivity of rice in later stages, ensures proper rice grain filling, and increases both grain filling percentage and grain weight, ultimately leading to higher yields [26,27]. In this study, the highest yield was achieved with a planting density of 30 cm × 13.3 cm when the base fertilizer, tillering fertilizer, and panicle fertilizer were applied in a ratio of 5:3:2. Under these conditions, both the number of panicles m⁻² and grain weight were optimal among the eight treatments. De Datta believed that [28] the application of N fertilizer during the early stages of rice growth primarily affected the number of tillers. In contrast, applying N fertilizer in the later stages mainly impacted the number of grains per panicle and the grain filling percentage. The distribution ratio of N fertilizer essentially reflects the crop’s varying demand for nitrogen at different growth stages. Under the conditions of this experiment, reducing the proportion of base fertilizer and increasing the proportion of ear fertilizer application more effectively increases yield. Therefore, with the same N application rate, decreasing the proportion of base fertilizer while increasing that of panicle fertilizer effectively promotes rice tillering and the formation and maturity of rice grains. The ratio of N fertilizer applied at different rice growth stages also has a notable impact on N use efficiency. Zhao et al. found that delaying the application of N fertilizer promoted deeper rooting of the main root, which increased root bleeding volume. This N fertilizer management approach led to strong root activity in rice, providing adequate nutrients and material support for the grain during the grain-filling stage, promoting filling, and ultimately increasing yield [29]. A balanced fertilization treatment with a 5:3:2 ratio of base, tillering, and panicle fertilizers can meet the diverse nitrogen requirements during the seedling, tillering, jointing, booting, and flowering stages of rice. This method also reduces N loss caused by overfertilization and improves N fertilizer absorption efficiency [30,31]. This aligns with our research findings. When the nitrogen ratio in base fertilizer, tillering fertilizer, and panicle fertilizer is set at 5:3:2, a higher yield can be achieved compared to other fertilization ratios. The increase in yield may be attributed to improved root traits. According to the research, enhancing the root traits of a single plant at the heading stage benefits nitrogen absorption and, consequently, yield. An appropriate N fertilizer ratio also contributes to the improvement of root bleeding [32].

N application rate also played a crucial role in rice yield. In this study, when the N application rate increased from 125 kg hm⁻² to 150 and 175 kg hm⁻², the yield increased by 9.9% and 6.6%, respectively. This aligns with the results of previous studies [33]. Increasing yield capacity is essential for achieving high overall yields. The varying responses of rice grain yield to N application rates and planting densities can mainly be attributed to the number of panicles per unit area and the number of spikelets per panicle. In this study, the impact of the N application rate and planting density on the number of panicles was greater than their effect on the number of spikelets per panicle. N application mainly increased the number of rice grains by promoting tillering. By contrast, planting density mainly increased the number of rice grains by increasing the number of basic seedlings. Increasing N application significantly increased rice yield and yield components [34]. According to the study, as the N application rate rises, both the light interception rate and intercepted radiation before and after flowering increase. As a result of higher N application rates, there was an increase in the number of rice tillers, the leaves, the leaf area index, and the capacity for light interception [35]. However, an excessive application rate of N can lead to redundant growth of rice roots. In contrast, an appropriate N application rate helps avoid this unnecessary root growth and facilitates greater transport of nitrogen to the shoots while ensuring a higher yield of rice. This results in both high yield and efficient utilization of nitrogen fertilizer [36]. Therefore, reasonably increasing the density of N fertilizer, configuring the proportion of N fertilizer, and reducing N application can increase the number of panicles per unit area, thereby increasing rice yield.

4.2. Effects of Different Planting Density and N Application Rate on N Use Efficiency and Photosynthetic Rate

From the perspective of N absorption and utilization, rice yield can be expressed in terms of a product of total plant N absorption and grain nitrogen production efficiency (NUEg) at maturity [37]. The indexes of N fertilizer absorption, absorption and utilization rate, agronomic utilization rate, physiological utilization rate, and partial productivity provide different insights into the absorption and utilization of N fertilizer in rice. Improving the utilization rate of N fertilizer is one of the main research directions for efficient and high-yield rice production. In this study, the total accumulation of N increased with higher application rates of N fertilizer. Under the same density treatment, both the agronomic utilization rate and the apparent utilization rate of N fertilizer were highest in the N2 treatment. In high-density treatment conditions, the 150 N treatment exhibited a higher N harvest index, while the 175 N treatment had a higher N harvest index under low-density conditions. An increase in rice tillers at elevated planting densities may explain this, as it enhances the number of panicles per square meter and mitigates yield losses associated with lower N application. While the productivity of N fertilizer and the physiological rate of N utilization rose with increased planting densities, they declined when N application rates were elevated. In comparison to other treatments, the D1N2 treatment exhibited a greater total accumulation of N, a higher apparent utilization rate of N, and an improved N harvest index.

From the perspective of SPAD value, leaves are the main source of rice, and dry matter is mainly produced through leaf photosynthesis [38]. Appropriate fertilization and density conditions can enhance the micro-meteorological environment within the rice population, thereby increasing the SPAD value of rice by extending the duration of photosynthesis in the rice canopy [39]. Our results indicated that the D1N2 treatment exhibited higher SPAD values at both HD and FS. Increasing planting density while reducing N input could increase SPAD values in the late rice growth stage, contributing to its high yield.

From the perspective of the net photosynthetic rate, photosynthesis plays a crucial role in material accumulation in crops, with the net photosynthetic rate determining the rate of this accumulation [40]. We found that the D1N2 treatment resulted in a higher net photosynthetic rate at both the PI and FS stages. Increasing planting density with reduced N input could lead to greater dry matter accumulation by enhancing the net photosynthetic rate during these stages.

4.3. Effects of Different Planting Density and N Application Rate on Roots

The root system plays a crucial role in absorbing and transmitting water and nutrients, and the above-ground parts of the supporting plants. The branching status and configuration of the root system are key factors influencing nutrient absorption efficiency [41]. It has been reported that planting density is an important variable impacting the total number of roots, total root length, root count per plant, total root length, and root dry weight in rice populations [42]. However, some studies have indicated that root length, root dry weight, and root volume at different growth stages are not significantly influenced by density treatments, which aligns with our research findings. This may be attributed to the minimal increase in our planting density. This increase still maintains the minimum area required for rice growth and provides adequate space and nutrients for its normal development, resulting in no significant effect on the root system. Increasing N fertilizer application can significantly enhance soil fertility, improve root physiological function, and increase root dry weight and length, facilitating crop absorption of water and nutrients. However, excessive N application inhibits root elongation and distribution [43,44]. The findings from this research demonstrated that, with the rise in N fertilizer application, the root length, dry weight of the roots, and root volume of rice considerably increased at every growth stage. Interestingly, the root length, root dry weight, and root volume observed in the N reduction treatment were marginally greater compared to the high N treatment. This suggests that a suitable decrease in the N application rate positively influences root development.

Our study demonstrated that increasing planting density under reduced N input mainly increased the root activity of rice during the PI and full HD stages, promoting the transport of nutrients from the underground parts to the panicle. The level of root oxidation power directly reflects the strength of rice root physiological activity. Higher root oxidation power improves the ability of rice to absorb fertilizers and water resources from the soil [45]. In this study, the increase in root activity may be attributed to the optimized cultivation method, which significantly reduced the root-to-shoot ratio and minimized ineffective root growth. This optimization also facilitated greater transport of fertilizers and water to the plant, ultimately increasing the yield. Nitrate plays a crucial role in enabling rice roots to absorb nitrogen from the soil. NR plays a key role in nitrate absorption [46]. GS acts as an important physiological index for measuring N use efficiency in plants [47]. We found no significant difference in GS and NR enzyme activity in rice roots across different density treatments. Increasing the N application rate enhanced the activities of both GS and NR enzymes. Moreover, higher planting densities under reduced N input could maintain higher GS and NR activity during the FS.

5. Conclusions

Under the condition of increasing planting density, using a fertilizer application ratio of 5:3:2 can enhance yield by increasing the panicles number per unit area and the grain weight. Compared to the traditional cultivation methods used by farmers, increasing planting density while reducing N input can effectively increase the panicles number per unit area of rice and boost overall yield. In addition, this approach increases total N accumulation, AEN, REN, NHI, SPAD value, photosynthetic rate, root length, root volume, root dry weight, root activity, and the activity of GS and NR enzymes. Meanwhile, the utilization efficiency of N fertilizer resources, photosynthetic rate, root physical properties, and root activity of rice were significantly improved. This enhancement in root absorption capacity, N fertilizer utilization efficiency, and photosynthetic rate contributed to increased yield. It is evident that an appropriate proportion of N fertilizer in the central region of Jilin Province, combined with increased planting density while reducing N input, can synergistically enhance both yield and N use efficiency. This approach promotes the development of resource-saving and eco-friendly cultivation techniques.